Legal claims defining the scope of protection, as filed with the USPTO.
1. A computer-implemented method of performing processor pipeline management, comprising: maintaining a registry of all ticket IDs, wherein each entry of the registry includes one of the ticket IDs; receiving, using a processor, an instruction for processing; determining, using the processor, that data in a first memory sub-group of a memory group needed to process the instruction is not available in a cache that ensures fixed latency access; determining, using the processor, that the instruction should be put in a sleep state based on the data in the first memory sub-group of the memory group not being available in the cache, wherein the sleep state indicates that the instruction will not be reissued until the instruction is moved to a wakeup state; determining an order in which data from all memory sub-groups, including the first memory sub-group and a second memory sub-group, of the memory group will be added to the cache; associating, using the processor, the instruction with a next-best ticket identifier (ID), wherein the next-best ticket ID corresponds, in the registry, with the second memory sub-group of the memory group rather than with the first memory sub-group of the memory group needed by the instruction, wherein the associating the next-best ticket ID includes determining that the second memory sub-group is a closest preceding one to the first memory sub-group in the order among all memory sub-groups of the memory group that correspond with any ticket ID in the registry; and moving, using the processor, the instruction to the wakeup state based on data associated with the next-best ticket ID, which is the data in the second memory sub-group of the memory group rather than the data in the first memory sub-group of the memory group, being moved into the cache.
2. The computer-implemented method according to claim 1 , wherein the memory group is a first memory group and the next-best ticket ID is a first ticket ID, the method further comprising maintaining the registry of all ticket IDs including the first ticket ID and one or more other ticket IDs, wherein each of the one or more other ticket IDs corresponds, in the registry, with the first memory group and a different memory sub-group than the first memory sub-group or the second memory sub-group or with a different memory group than the first memory group and a memory sub-group of the different memory group.
3. The computer-implemented method according to claim 2 , wherein the associating the instruction with the first ticket ID includes selecting the first ticket ID from the registry.
4. The computer-implemented method according to claim 2 , wherein the maintaining the registry of all ticket IDs includes maintaining a catch-all ticket ID whose entry in the registry does not include any particular memory group.
5. A system to perform processor pipeline management, comprising: a cache that provides fixed-latency storage access; and a processor configured to maintain a registry of all ticket IDs, wherein each entry of the registry includes one of the ticket IDs, to obtain an instruction for processing, to determine that data in a first memory sub-group of a memory group needed to process the instruction is not available in the cache, to determine that the instruction should be put in a sleep state based on the data in the first memory sub-group of the memory group not being available in the cache, wherein the sleep state indicates that the instruction will not be reissued until the instruction is moved to a wakeup state, to determine an order in which data from all memory sub-groups, including the first memory sub-group and a second memory sub-group, of the memory group will be added to the cache, to associate the instruction with a next-best ticket identifier (ID), wherein the next-best ticket ID corresponds, in the registry, with the second memory sub-group of the memory group rather than with the first memory sub-group of the memory group needed by the instruction and the processor is configured to associate the next-best ticket ID by determining that the second memory sub-group is a closest preceding one to the first memory sub-group in the order among all memory sub-groups of the memory group that correspond with any ticket ID in the registry, and to move the instruction to the wakeup state based on data associated with the next-best ticket ID, which is the data in the second memory sub-group of the memory group rather than the data in the first memory sub-group of the memory group, being moved into the cache.
6. The system according to claim 5 , wherein the memory group is a first memory group and the next-best ticket ID is a first ticket ID, wherein the registry of all ticket IDs includes the first ticket ID and one or more other ticket IDs, and each of the one or more other ticket IDs corresponds, in the registry, with the first memory group and a different memory sub-group than the first memory sub-group or the second memory sub-group or with a different memory group than the first memory group and a memory sub-group of the different memory group.
7. The system according to claim 6 , wherein the processor is configured to associate the instruction with the first ticket ID by selecting the first ticket ID from the registry.
8. The system according to claim 6 , wherein the registry of all ticket IDs includes a catch-all ticket ID whose entry in the registry does not include any particular memory group.
9. A computer program product for performing processor pipeline management, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to perform a method comprising: maintaining a registry of all ticket IDs, wherein each entry of the registry includes one of the ticket IDs; receiving an instruction for processing; determining that data in a first memory sub-group of a memory group needed to process the instruction is not available in a cache that ensures fixed latency access; determining that the instruction should be put in a sleep state based on the data in the first memory sub-group of the memory group not being available in the cache, wherein the sleep state indicates that the instruction will not be reissued until the instruction is moved to a wakeup state; determining an order in which data from all memory sub-groups, including the first memory sub-group and a second memory sub-group, of the memory group will be added to the cache; associating the instruction with a next-best ticket identifier (ID), wherein the next-best ticket ID corresponds, in the registry, with the second memory sub-group of the memory group rather than with the first memory sub-group of the memory group needed by the instruction, wherein the associating the next-best ticket ID includes determining that the second memory sub-group is a closest preceding one to the first memory sub-group in the order among all memory sub-groups of the first memory group that correspond with any ticket ID in the registry; and moving, using the processor, the instruction to the wakeup state based on data associated with the next-best ticket ID, which is the data in the second memory sub-group of the memory group rather than the data in the first memory sub-group of the memory group, being moved into the cache.
10. The computer program product according to claim 9 , wherein the memory group is a first memory group and the next-best ticket ID is a first ticket ID, the method further comprising maintaining the registry of all ticket IDs including the first ticket ID and one or more other ticket IDs, wherein each of the one or more other ticket IDs corresponds, in the registry, with the first memory group and a different memory sub-group than the first memory sub-group or the second memory sub-group or with a different memory group than the first memory group and a memory sub-group of the different memory group.
11. The computer program product according to claim 10 , wherein the associating the instruction with the first ticket ID includes selecting the first ticket ID from the registry.
Unknown
October 19, 2021
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.